Root dynamic growth strategies in response to salinity

Abstract Increasing soil salinization largely impacts crop yield worldwide. To deal with salinity stress, plants exhibit an array of responses, including root system architecture remodelling. Here, we review recent progress in physiological, developmental and cellular mechanisms of root growth responses to salinity. Most recent research in modulation of root branching, root tropisms, as well as in root cell wall modifications under salinity stress, is discussed in the context of the contribution of these responses to overall plant performance. We highlight the power of natural variation approaches revealing novel potential pathways responsible for differences in root salt stress responses. Together, these new findings promote our understanding of how salt shapes the root phenotype, which may provide potential avenues for engineering crops with better yield and survival in saline soils.


| INTRODUCTION
Increased soil salinity is a worldwide problem that causes yield loss of crops. Due to the current rising sodium chloride levels in groundwater, secondary salinization affects irrigated land, eventually leading to the loss of agricultural soils (FAO, 2015). Unlike halophyte species that are highly tolerant to salt concentrations (at least 200 mM NaCl), glycophyte plant species, that include most of our crops, can only grow healthily in low concentrations of salt (Cheeseman, 2015;Flowers & Colmer, 2008). Thus, research efforts aim to find physiological and genetic solutions to minimize the impact of saline land on global crop yield (Ismail & Horie, 2017;Munns et al., 2020;Munns & Gilliham, 2015). Salt stress is defined as the detrimental effect of high concentrations of salt accumulated in the soil, leading to inhibition of plant growth and development (Rahman, Ijaz, Qamar, Bukhari, & Malik, 2018). In saline soils, plants experience both osmotic and ionic stress. Osmotic stress arises due to the increased sodium ions in the soil, which leads to the reduction of water absorption, affecting various downstream processes in plants within several hours (Awlia, Alshareef, Saber, et al., 2021;Julkowska & Testerink, 2015). The ionic stress component was believed to affect plants much later after perception of the Na + stimulus; when toxic NaCl levels were reached in the root and shoot tissues (Munns & Tester, 2008). However, recent studies show that plants can generate rapid ionic stress-specific Ca 2+ signals within 30 s (Choi, Toyota, Kim, Hilleary, & Gilroy, 2014;Jiang et al., 2019) and show internalization of PIN2 auxin transporters within an hour as a specific response to Na + ions (Galvan-Ampudia et al., 2013). Cellular signal transduction pathways induced upon salinity include calcium and cyclic guanosine monophosphate signalling, phospholipid signalling and reactive oxygen species (ROS) formation as well as protein kinase activation (Lamers, van Meer, & Testerink, 2020;van Zelm, Zhang, & Testerink, 2020). Moreover, cellular Na + /K + balance modulated via ion transporters such as potassium channels and (anti-) transporters also contribute to plant local and distal salt responses  Y. Yang & Guo, 2018). Together, these cellular processes affect plant growth and survival in salt.
Roots are able to sense multiple environmental stimuli, coordinate cellular stress responses and reorient their growth direction as reviewed Muthert, Izzo, van Zanten, & Aronne, 2020). Root growth strategies are dynamically changed in response to biotic and abiotic stresses, by modulation of specific key traits, such as root length and branching, redirection of root growth and modification of cell wall compositions. In salinized conditions, plants exhibit root phenotypic plasticity by modulating both root system architecture (RSA) components and directional growth dynamically (Dinneny, 2019;Julkowska et al., 2017;Korver et al., 2020). Understanding these dynamic root growth strategies and their underlying physiological, developmental and cellular mechanisms is expected to contribute to future agricultural strategies to improve crop yield in saline soils.
This review provides an overview of root architectural plasticity in responses to salinity stress and links new findings on mechanisms to phenotypes. We firstly discuss the basics of root phenotypes under default conditions without salt, and then compare to salinity conditions focusing on the physiological responses, root developmental growth, genetic and natural variation approaches and cell wall modulation.
2 | RSA REMODELING STRATEGIES TO COPE WITH SALINITY 2.1 | RSA: basic strategies to shape the root The soil environment is variable and complex, thus to grow in soil, the development of roots needs to be flexible in response to many cues (Koevoets, Venema, Elzenga, & Testerink, 2016;Rellán-Alvarez, Lobet, & Dinneny, 2016). The RSA of a plant describes the spatial configuration of the whole root system, that exhibits a considerable diversity among species, genotypes, space and time (Lynch, 1995). Monocotyledons and dicotyledons show fundamentally distinct patterns in RSA, as reviewed (Osmont, Sibout, & Hardtke, 2007). In general, RSA, as a plastic trait, presents the root 3D developmental plasticity that is established to avoid unfavourable environments and optimize the utilization of resources (Morris et al., 2017

| Salt modulates RSA
Plants exhibit notable changes in RSA in response to salt. The effect of salt treatment on root growth is largely dependent on the severity of the salt treatment. In wheat, salt inhibits root length in a dosedependent manner ranging from 50 to 200 mM NaCl (Rahnama, Munns, Poustini, & Watt, 2011). In Arabidopsis, moderate to high salt concentrations (75-150 mM NaCl) inhibit both primary root (PR) and lateral root (LR) growth ( Figure 1). The traits of PR length, LR length and LR number are consistently reduced, but LR density shows a large variation in response to high salt concentrations between different reports, likely caused by different nutrient concentrations used (Julkowska et al., 2014;P. Li, Yang, et al., 2021;Y. Zhao, Wang, Zhang, & Li, 2011;Zolla, Heimer, & Barak, 2010).
Under even higher salt concentrations (NaCl ≥ 200 mM), the emerging and young LRs (<100 μm) exhibit dramatically less damage, as compared with the PR and elongated LRs (> 400 μm) in viability assays (Ambastha, Friedmann, & Leshem, 2020). The difference observed could be due to the higher level of NADPH oxidaseactivated ROS induction in the young LR as compared to the PR (Ambastha et al., 2020). On the other hand, the relevance of these observations remains to be established, as they occur on a salt concentration that is lethal to the plants. Interestingly, low salt concentrations (<NaCl 50 mM) promote LR growth when 4 days or 5 days-old seedlings are transferred to the NaCl stress medium (Julkowska et al., 2014;Zolla et al., 2010). Additionally, this effect was shown to be specific to salt stress, and did not occur in response to osmotic stress (Zolla et al., 2010). When seeds are germinated directly on the low-stress medium of 30 mM NaCl, no difference in ). When exposed to salt, the number of root hairs is dramatically reduced, which could be partially attributed to the large decrease in epidermal cell numbers under salt stress (Dinneny, Long, Wang, et al., 2008;. Notably, ionic stresses (NaCl, KCl or LiCl) could reduce both root hair density and elongation, while an equi-osmolar concentration of mannitol increased both root hair density and elongation, indicating different mechanisms in regulating root hair growth between ionic and osmotic stresses (Y. Wang et al., 2008). In response to salt treatment, both cell autonomous and non-autonomous effects were observed by transcriptional profiling of three root hair epidermal patterning mutants.
Among the identified salt-responsive genes, many of these genes repressed by salt were involved in cell wall structure and trichoblast differentiation (Dinneny et al., 2008). Interestingly, salt inhibited the root hair outgrowth immediately after the treatment but then resumed after 8 hr, indicating the root hair responses are dynamic (Dinneny et al., 2008). However, current knowledge is still very limited on root hair developmental regulation and its contribution to salt stress responses.  (Abas et al., 2006;Brunoud et al., 2012;Fendrych et al., 2014;Fendrych, Leung, & Friml, 2016). Auxin redistribution largely depends F I G U R E PIN3, PIN4 and PIN7-work together to establish a root "auxin reflux loop" as summarized in reviews (Adamowski & Friml, 2015;Korver, Koevoets, & Testerink, 2018;Zhou & Luo, 2018).
Under high salinity, plants can reduce their exposure to salinity by changing their roots' growth direction to avoid a saline environment through a response called halotropism. Not only the roots of Arabidopsis, but also the roots of tomato (Solanum lycopersicum) and sorghum (Sorghum bicolor) seedlings, either on agar media or in soil, are able to exhibit a halotropic response when exposed to a salt gradient (Galvan-Ampudia et al., 2013). Halotropism is accomplished by redistribution of auxin in the root tip mainly by the PIN2 auxin efflux carrier. When exposed to a salt gradient, salt triggers internalization of PIN2 at the saltier side of the root, consequently leading to auxin redistribution and the directional bending away from salt (Galvan- Ampudia et al., 2013). In addition, other auxin transporters are also relevant to halotropism. For example, a recent study showed that salt induces the transient upregulation of PIN1 in the stele, and an elevation of AUX1 levels on the non-salt-exposed side (van den Berg, Korver, Testerink, & ten Tusscher, 2016). In accordance, pin1 mutants show a delayed halotropic response, demonstrating a role for PIN1 in halotropism (van den Berg et al., 2016; Figure 2).
Notably, it was shown that osmotic stresses (both NaCl and mannitol treatments), immediately enhance clathrin-mediated endocytosis, increase PIN endocytosis and overcome the inhibitory effect of auxin (Baral et al., 2015;Nakayama et al., 2012;Zwiewka, Nodzy nski, Robert, Vanneste, & Friml, 2015). Moreover, a recent study showed that both NaCl and sorbitol treatments induce large membrane structures at the plasma membrane in epidermal and LR cap cells (Korver  , 2020). In response to both ionic and osmotic stresses, there is a crucial role for ζ-type PLD (PLDζ1) that is required for the relocalization of auxin carrier PIN2, one of the major factors in halotropism (Korver et al., 2020). On the other hand, despite the overlap with osmotic responses, roots also seem to exhibit a NaCl-specific cellular response as a NaCl gradient affected PIN2 subcellular localization in epidermal cells while mannitol did not affect PIN2 localization, and also the observed phenotypic response is specific to NaCl and not

| UNCOVERING THE GENETIC CONTROL OF ROOT PLASTICITY
Natural variation studies have proven to be successful in identifying novel genetic components and uncovering relevant loci that are associated with agronomic traits. This approach relates phenotypic variation to genetic variation thereby statistically associating traits with sequence polymorphism in natural accessions (Alonso-Blanco et al., 2016;Seren et al., 2013). Several novel root growth regulation factors identified via genome wide association studies (GWAS) were reported to be associated with RSA traits in both default (without salt) and saline conditions. In Arabidopsis, Cytokinin oxidase 2 (CKX2) was identified to be associated with a gravitropic setpoint angle of LRs.
CKX2 affects LR directional growth and cellular elongation by controlling the metabolism of cytokinin (Waidmann, Ruiz Rosquete, Schöller, et al., 2019). Additionally, Exocyst70A3 was identified to control the depth of the root system by modulating PIN4-dependent auxin transport (Ogura et al., 2019). Both CKX2 and EXO70A3 modulate RSA For crops, genetic control of RSA is largely unelucidated, while natural variation in RSA plasticity in response to salt has been found in maize and rice. In maize, the Aux/IAA TF family member ZmIAA1 and the GRAS-type TF family member ZmGRAS43, were identified to be associated with modulating RSA in response to salinity stress (P. Li, Yang, et al., 2021). Another recent study showed that rice yields in saline soil can be improved via a QTL containing SOIL SUR-FACE ROOTING 1 (qSOR1), a homolog of DEEPER ROOTING 1 (DRO1), which is modifying root growth direction to a shallower root growth angle . However, in the absence of ion accumulation data of the soil the physiological relevance of the benefit of developing shallower root systems in this case is still unclear. In contrast, tomato seedlings growing in rhizotrons placed their LRs preferentially lower, avoiding the higher soil layers in which salt accumulated (Gandullo et al., 2021). Here, the genetic loci that contribute to the response remain to be characterized. Thus, a better understanding of RSA modulation is required to directly contribute to salt tolerance and may provide guidance to future breeding in salinized soil.  (Byrt, Munns, Burton, Gilliham, & Wege, 2018). Cell wall composition can be changed dynamically in response to biotic and abiotic stresses, and differs between species and cell types (Feng, Lindner, Robbins, & Dinneny, 2016;Vaahtera, Schulz, & Hamann, 2019;Wolf, Hématy, & Höfte, 2012).
Recent studies show that biosynthesis genes of cell wall components, including lignin (A. Q. Duan et al., 2020), arabinose (C. Zhao et al., 2019) and galactan (Yan et al., 2021), can contribute to salt tolerance. The modulation of root structural barriers via cell wall composition changes could reduce water loss and limit Na + entry in roots (Byrt et al., 2018). Such changes include lignification and suberization in both endodermis and exodermis (Barberon, Vermeer, De Bellis, et al., 2016;Kajala et al., 2021). Lignification in higher plants provides firmness and hydrophobicity to the secondary cell walls and thus forms a barrier to protect plant against stresses (Q. Zhao, 2016). Similarly, suberization also generates hydrophobic and lipophilic secondary cell wall macromolecules acting as a barrier to restrict water and solutes (Andersen, Barberon, & Geldner, 2015). Salt stress increases root lignin content while reducing arabinoxylan content in maize (Oliveira et al., 2020). In Arabidopsis, overexpression of a suberin biosynthesis enzyme β-Ketoacyl-CoA Synthase (VvKCS) from grape Vitis vinifera L.
The Casparian strip, primarily made of lignin, is affecting the radial transport of water and minerals to the vasculature tissue (Naseer et al., 2012). After exposure to salt, the radial width of the Casparian strip was increased in maize (Karahara, Ikeda, Kondo, & Uetake, 2004). In Arabidopsis, salt stress enhances suberization of the root endodermal layer (Barberon et al., 2016), while in rice and maize lignification and suberization occur at both endodermal and exodermal root layers (Krishnamurthy et al., 2009

| CONCLUSION AND PROSPECTS
Plants can dynamically adjust their root growth components (e.g., density, length and angle) locally in response to salinity, to minimize their metabolic cost and the impact of the stresses. Local root growth plasticity responses appear crucial for plants to optimize architecture spatially and temporally under stress conditions (Julkowska et al., 2017;Kazan & Lyons, 2016;Korver et al., 2018). However, the regulation of salt-induced root growth plasticity is still largely unknown. The current available tools using plant tissue-specific and cell-specific approaches that can analyse a few cells or a small area specifically based on mass spectrometry, biosensors or single-cell omics can help us to unravel the mechanisms underlying local root growth in response to salt stress (Chen et al., 2020;Novák, Napier, & Ljung, 2017). We speculate that by activating local stress response pathways to accomplish local growth plasticity, salt stress could induce not only local developmental changes, but also systemic or distal signals, which are also crucial for adequate stress responses (H. Li, Testerink, et al., 2021).
This review discussed the regulation of dynamic root growth by salt and the underlying mechanisms that are expected to contribute to future agricultural efforts to promote crop stress resilience. In a natural heterogeneous soil, environment roots need to flexibly respond to several environmental cues. Thus, plant root growth plasticity in saline soil presents the outcome of the root dynamic growth affected by salt stress, soil type, climate, biota and other environmental conditions.
Further prediction and modelling for the complex interactions between roots and saline soil is required, to provide insights into the mechanisms and new leads for improving agricultural production.
Moreover, more research efforts are needed to expand the knowledge beyond Arabidopsis; the lack of molecular tools is hindering the study of different crop species and natural varieties. On the positive side, natural variation approaches have successfully identified new genetic components contributing to salinity tolerance in crops species, including rice (Patishtan, Hartley, de Carvalho, & Maathuis, 2018), maize (M. Zhang, Liang, et al., 2019), tomato (Z. Wang et al., 2020), soybean (W. Zhang, Liao, et al., 2019) and barley (Hordeum vulgare; Hazzouri et al., 2018), which will be valuable molecular entries to further characterize the molecular mechanisms and to contribute to future engineering of more resilient crops.
Finally, the progress in developing new phenotyping techniques and platforms will help us capture the diverse root local responses among different species in both spatial and temporal aspects, and their relevance for salt resilience. For example, different RSA patterns were shown to be correlated with shoot Na + /K + homeostasis in Arabidopsis when they were grown in agar plates (Julkowska et al., 2014). In tomato plants grown in soil, the most severe change in RSA under salt conditions was the suppression of LR emergence at the soil surface were salts accumulate, which could be an adaptive response (Gandullo et al., 2021). On the other hand, in rice, the near-isogenic line qsor1-NIL that had 'soil-surface roots' showed a significant increase in yield in saline paddies, compared to the cultivars without the soil-surface roots . Therefore, further evidence on a range of species and soil conditions is needed before general or specific beneficial root response strategies can be identified that would be future target phenotypes for breeding to optimally support whole plant salt resilience.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.

DATA AVAILABILITY STATEMENT
Data sharing not applicable to this article as no datasets were generated or analysed during the current study.